[pvrusb2] read attempt on video device leads to error

[Frans Meulenbroeks]

Lars, Mike,

Just read the whole mail thread (I'm getting the
digest).
Secam can be outruled completely. I have a secam unit
and  indeed then secam is the default. I've used it
that way and you won't get proper color (and perhaps
no audio), but it cannot cause this problem. That is,
in secam things work for me.

The PCF8574 is a completely different issue. This is
an i2c I/O expander.

The datasheet of the TI version is here:
http://focus.ti.com/lit/ds/symlink/pcf8574.pdf
This one says the i2c address of it is defined by hw
lines and indeed it can be 21 hex.

The Philips page of the device is:
http://www.semiconductors.philips.com/pip/PCF8574.html
with a datasheet on:
http://www.semiconductors.philips.com/acrobat_download/datasheets/PCF8574_4.pdf

Lars, is there any indication (e.g. in the log or so)
that you have such a device in your PC?

When you connect a pvrusb a new i2c bus is registered.
I have peeked into the code and all i2c drivers are
triggered to check if there is a device they can
handle on this new bus. A badly written driver for the
8574 could see a device at address 21 and tell that it
can handle it. If that is really the case you might
need to disable or improve the 8574 code. 

I fear that you'll have to recompile the kernel with
some debug printk's in it to see why the 8574 driver
latches on onto the 7115 address.

Good luck, Frans.

(ps, if you want to discuss this further, when I am
online in the evenings or weekend (CET), i try to
hangout on #pvrusb2 on freenode (nick eFfeM), feel
free to page me there).



		
__________________________________ 
Yahoo! Music Unlimited 
Access over 1 million songs. Try it free.
http://music.yahoo.com/unlimited/